Backup detected log corruption in a client’s database, and every full backup failed at exactly 60 percent. Instead of rebuilding the log, we checked the VLFs and found a log waiting on REPLICATION. Removing the leftover replication fixed the backup.
CPerIndexMetaQS::ErrorAbort: Too Many Dumps From Index Corruption
CPerIndexMetaQS::ErrorAbort in a stack dump points to index corruption in one of your databases. A reader had too many dumps, and the ERRORLOG and DBCC CHECKTABLE showed which index was bad. Here are the two repair options and why backups matter.
SQL SERVER – Msg 0, Level 11 – A Severe Error Occurred on the Current Command. The results, if Any, Should be Discarded
Msg 0, Level 11 says a severe error occurred but gives no clue how to fix it. Here is my checklist, starting with DBCC CHECKDB on the database and table.
SQL SERVER – 5 Don’ts When Database Corruption is Detected
Here is an email I received about database corruption and that has lead me to write this quick blog post about database corruption.
SQL SERVER – Running CHECKDB with Minimum Repair Level
CHECKDB found 39 consistency errors and said repair_allow_data_loss is the Minimum Repair Level, but the reader could not finish the repair. Here is my reply.
Error 9003: Log Scan Number in master Is Not Valid
Error 9003 on the master database means SQL Server cannot recover master and will not start. The master files were corrupt. We rebuilt the system databases, started the instance in single user mode and restored master from a backup.
SQL Agent Alerts for Severity 17 to 25 and Errors 823, 824, 825
Configure Agent alerts for serious severities and errors 823, 824 and 825, connect an operator to Database Mail, and test response delays.